GadaaLabsGadaaLabsCommunityยท24 Agent Skills Included

claude-code-on-steroids

Structured coding workflows, debugging discipline, and token cost analytics

Enforces disciplined software workflows: test-driven development, root-cause debugging, code review gates, and verified completion before any success claim. Cuts wasted tokens with memory caching, context compression, model routing, and a built-in usage dashboard. Adds domain playbooks for ML, frontend, embedded, and LLM systems, plus parallel subagent coordination for multi-file tasks.
npx skills add GadaaLabs/claude-code-on-steroids --all -g -y
Available:

Tells the AI agent how the 24 skills are organized, how to safely modify behavior-enforcing skill files, and how codenames map to workflows before making any change.

All Skills in This Repository (24)

Pure Emerald Level Indicators
๐Ÿ“ฆ In Repo
GadaaLabsGadaaLabs

hunter

Investigate root causes through a four-phase debugging workflow before implementing fixes.

Community
Advanced
๐Ÿ“ฆ In Repo
GadaaLabsGadaaLabs

gradient

Apply ML engineering patterns to data pipelines, training, serving, and MLOps.

Community
Advanced
๐Ÿ“ฆ In Repo
GadaaLabsGadaaLabs

commander

Dispatch parallel agents with isolated contexts for independent tasks.

Community
Advanced
๐Ÿ“ฆ In Repo
GadaaLabsGadaaLabs

ascend

Enforces a skill invocation protocol at the start of every session.

Community
Advanced
๐Ÿ“ฆ In Repo
GadaaLabsGadaaLabs

pathfinder

Map an unknown codebase through a structured five-phase audit.

Community
Intermediate
๐Ÿ“ฆ In Repo
GadaaLabsGadaaLabs

oracle

Classify task complexity across 8 dimensions and select an execution chain.

Community
Advanced
๐Ÿ“ฆ In Repo
GadaaLabsGadaaLabs

prism

Identifies UI/UX problems and proposes domain-guided solutions for web/app development.

Community
Advanced
๐Ÿ“ฆ In Repo
GadaaLabsGadaaLabs

seal

Finalize feature branches by verifying tests and presenting four completion options.

Community
Advanced
๐Ÿ“ฆ In Repo
GadaaLabsGadaaLabs

horizon

Monitor context health and compress stale reasoning during long sessions.

Community
Advanced
๐Ÿ“ฆ In Repo
GadaaLabsGadaaLabs

sentinel

Enforce fresh verification evidence before accepting completion claims in software development.

Community
Advanced
๐Ÿ“ฆ In Repo
GadaaLabsGadaaLabs

exodus

Execute planned tasks in a fresh isolated session with strict checkpointing.

Community
Advanced
๐Ÿ“ฆ In Repo
GadaaLabsGadaaLabs

sculptor

Enforce test-driven development with a failing baseline test before skill implementation.

Community
Advanced

Frequently Asked Questions

FAQPage Schema
How to install claude-code-on-steroids?โ–ผ

Run `npx skills add GadaaLabs/claude-code-on-steroids --all -g -y` in your terminal to install all skills globally.

How to reduce token usage in Claude Code?โ–ผ

This suite caches past solutions in memory, compresses long-session context, routes simple tasks to cheaper models, and includes a /tokenburn dashboard that shows exactly where your tokens go.

How to stop AI from making random bug fixes?โ–ผ

The hunter skill forces root-cause investigation before any fix, and the sentinel skill blocks completion claims until tests are actually run and verified.

Does it work with Codex, Copilot, and Gemini CLI?โ–ผ

Yes. The ascend skill includes tool-mapping references for Codex, Copilot CLI, and Gemini CLI so the same skills run across platforms.

Can I use it without prompt engineering experience?โ–ผ

Yes. Skills trigger automatically from plain-English requests, and the oracle skill classifies each task and picks the right workflow for you.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’